Electrotyping Daguerreotypes: Reconstruction of an Early Reproduction Technique

(1)

Conservation and Restoration programme, Master’s, Photograph Specialisation

Electrotyping Daguerreotypes:

Reconstruction of an Early Reproduction Technique

Student:

Magdalena Pilko

Student number:

10666664

Supervisor:

Clara von Waldthausen, UvA Amsterdam

External advisors: Martin Jürgens, Rijksmuseum Amsterdam

Dr. Bill Wei, Dutch Cultural Heritage Agency

(2)

Acknowledgements

I am very grateful to Martin Jürgens, photograph conservator, Rijksmuseum Amsterdam, who kindly introduced me to his working methods and most generously provided me with unpublished information and other valuable resources at all times. My course coordinator, Clara von Waldthausen, UvA, I thank for her kind support in many ways, in particular for making materials available as well as for critical reflection and encouragement. I am very grateful to Dr. Bill Wei, RCE, who kindly advised me throughout the process of writing and to Dr. René Peschar, UvA, for his kind contribution of chemical expertise.

Furthermore, I would like to thank all UvA R&C Master / PI students of the metal department, who generously shared their workspace and tools, particularly Michaela Groeneveld and Marianne Nuij for their continued practical assistance. I thank Tonny Beentjes, coordinator of the metal department, UvA, who provided me with his personal rectifier and shared his practical knowledge in electrotyping. Thanks are also due to Tamar Davidowitz, metal conservator, UvA/Rijksmuseum Amsterdam, for her practical introduction into electrotyping.

I thank Nicholas Burnett, conservator, for his warm reception in Cambridge, not only providing me with a bicycle to commute to and from his laboratory but all of his facilities and by contributing data. I am grateful to Dr. Iris Buisman, University of Cambridge, and Dr. Ineke Joosten, RCE, for their patience in the search for image particles with SEM and to Bas van Velzen, coordinator of paper conservation, UvA, for assisting with the use of the Hirox microscope.

Thanks are also due to Anton Orlov, daguerreotypist, San Diego, USA, whom I experienced as a neighbour communicating to me on the making of his daguerreotypes in the most open and comprehensive manner, to Johan de Zoete, graphic craftsman, who provided me with literature and kindly offered further assistance in electrotyping, to Dr. Han Neevel, RCE, for his assistance with the interpretation of XRF and EDS results, to Marinus Ortelee, daguerreotypist, for offering the possibility to make daguerreotypes as well as to Rosina Herrera Garrido, Rijksmuseum Amsterdam, Prof. Dr. Marc Koper, professor for surface chemistry, Leiden University, Ellen van Bork, Rijksmuseum Amsterdam/UvA and Sanne Berbers, UvA, for their interest in my work. My fellow students, in particular, Kayleigh van der Gulik, I thank for sharing experiences.

Last but not least, I’m profoundly grateful to my family, Michael, Celia and Leonard who enabled me to do this work.

(3)

Abstract (English)

Electrotyping Daguerreotypes: Reconstruction of an Early Reproduction Technique, a thesis by

Magdalena Pilko in the context of the conservation and restoration master programme with a specialization in photography at the University of Amsterdam, 2017.

In the 19th_{century the electrotype process was used to reproduce daguerreotypes as copper plate} facsimiles. The characteristics of this technique can be studied best by examining the copper copy plate together with its master daguerreotype, but only a very small number of these plate pairs are known today. A reconstruction of the process was therefore attempted to understand whether the results from visual and analytical analysis of historical electrotyped daguerreotypes can be considered as typical characteristics of the process. For this reconstruction, Lerebours’ instructions from 1843 on electrotyping daguerreotypes, supplemented with information by other authors, were followed as precisely as possible. In addition to a literature study of historical and modern technical sources, three historical objects and reconstructions were photographically documented and examined visually in ambient and ultraviolet (UV) fluorescence light, with X-ray Fluorescence Spectroscopy (XRF) and in the Scanning Electron Microscope with Energy Dispersive Spectroscopy (SEM-EDS).

Working parameters for electrotyping daguerreotypes could be specified with a current density of approximately 3.46 A/dm2_{for approximately 8 hours resulting in an approximately 0.4 mm thick} electrodeposit. These parameters were applied to two contemporary and two 19th _century daguerreotypes. The initial results with visual examination and instrumental analysis with XRF and SEM-EDS indicates that during the separation of the plates, the top silver-gold layer including the image particles of the daguerreotype transferred from the daguerreotype to the electrotype. From this we may conclude that it is likely that some kind of surface treatment of the daguerreotype took place prior to electrotyping; however, this aspect is not described in the historical instructions considered in this study.

This author went beyond Lerebours’ instructions by applying a beeswax separation layer on one historical and two contemporary daguerreotypes prior to electrotyping them. This layer produced electrotypes that resemble the original historical objects by visual examination and by SEM microscopy. Although further research on the reconstruction is required, the results of this study as well as the modern daguerreotype electrotypes that were produced can be used to aid in the identification of yet unidentified daguerreotype electrotypes and in the study of their mechanisms of aging.

(4)

Abstract (Dutch)

Deze scriptie Electrotyping Daguerreotypes: Reconstruction of an Early Reproduction Technique is geschreven door Magdalena Pilko in het kader van de Masteropleiding in Conservering en Restauratie van Cultureel Erfgoed, specialisatie fotorestauratie, aan de Universiteit van Amsterdam, 2017.

In de 19e eeuw werden door galvanotechniek reproducties van daguerreotypieën op koperplaten gemaakt. Deze techniek, toegepast op daguerreotypieën, kan het beste bestudeerd worden aan de hand van een koperplaat kopie met zijn master daguerreotypie, maar er zijn maar weinig dergelijke sets van platen bekend. Daarom is er geprobeerd om dit procedé te reconstrueren, om zodoende te begrijpen of kenmerken van geanalyseerde historische galvano-daguerreotypieën met dit procedé geproduceerd zijn. De instructies van Lerebours over galvanoplastische reproducties van daguerreotypieën uit 1843 is aangevuld met informatie door andere auteurs en dat procedé is zo nauwkeurig mogelijk gereconstrueerd. Naast het bestuderen van historische en technische literatuur zijn drie historische objecten en resultaten van de reconstructies fotografisch gedocumenteerd en visueel onderzocht in zichtbaar licht en met ultraviolet (UV) fluorescentie, met X-ray Fluorescentie Spectroscopie (XRF) en met Scanning Electron Microscopie met Energy Dispersive Spectroscopy (SEM-EDS).

Werkparameters voor galvano-daguerreotypieën konden worden bepaald. Een stroomdichtheid van ongeveer 3,46 A/dm2 gedurende ongeveer acht uur, resultereerde in een ongeveer 0,4 mm dikke elektrodepositie. Deze parameters werden toegepast op twee hedendaagse en twee historische daguerreotypieën. Resultaten daarvan zijn niet vergelijkbaar met de onderzochte historische galvano-daguerreotypieën. Visueel onderzoek en analyse met XRF en SEM-EDS geven aan dat tijdens de scheiding van de platen de bovenste zilver-goud laag inclusief beelddeeltjes van het daguerreotype overgebracht wordt op de galvano-daguerreotypie. Hieruit kunnen wij concluderen dat het waarschijnlijk is dat er een soort behandeling van de daguerreotypie voorafgaand aan het galvaniseren plaatsvond, die niet in de gebruikte historische instructies is beschreven.

De auteur heeft Lerebours’ instructies uitgebreid en een scheidingslaag in de vorm van een bijenwaslaag op een historische en twee hedendaagse daguerreotypieën aangebracht, voorafgaand aan het galvaniseren. Dit heeft het mogelijk gemaakt galvano-daguerreotypieën te maken die volgens visueel en SEM-EDS onderzoek op de onderzochte historische objecten lijken. Hoewel verder reconstructieonderzoek nodig is, kan het resulterende onderzoek evenals de gereconstrueerde daguerreotypieën gebruikt worden bij de veilige identificatie van nog niet geïdentificeerde galvano-daguerreotypieën en bij de studie naar hun veroudering.

(5)

1. Introduction to the research project

1.1. Introduction

The daguerreotype process, publicly announced in 1839, was the first commercially successful photographic process. A daguerreotype has an extremely high image resolution due to image-forming silver-amalgam particles of submicron size on top of a silver-plated copper plate without the presence of a binder. Daguerreotypes were often gold-toned, which improved their optical qualities and their physical resistance.1_{The daguerreotype was very successful, but its major disadvantages were that the} photographic image was mirrored and difficult to reproduce. Several attempts were therefore made early on to find a method for reproducing daguerreotypes.2One method that proved successful was the electrotype process. Electrotyping, or electroforming, is the reproduction of a metal artefact by means of the electrodeposition of metal ions upon its conductive surface, resulting in a physical copy that is subsequently separated from the original artefact.3 In the 19th century, this process was widely applied to fabricate copper reproductions of various kinds of industrial articles and art objects such as sculptures or medals.4_{An electrotype taken from a daguerreotype depicts a mirrored image on a} copper plate. Figure 1 depicts a schematic diagram of the composite structure of a daguerreotype and its electrotype reproduction.

Figure 1. Schematic cross-section of a daguerreotype and its electrotype.5

Illustration: M. Pilko.

Although the reproductions of daguerreotypes obtained by electrotyping were praised for their fine results in 19th_{century sources such as in a treatise by the Austrian physician and photographer Anton} Martin, at present only 13 copper plates have been identified as probable electrotypes in different collections worldwide (see Appendix III – Overview of potential daguerreotype electrotypes in collections).6_{Little has been published on the process and its characteristics, and it is therefore} possible that unidentified daguerreotype electrotypes exist in collections.

To gain a better understanding of the technology and materials of the electrotype process as applied to daguerreotypes, the photograph conservator of the Rijksmuseum Amsterdam, The Netherlands, Martin Jürgens, in partnership with others, has been researching this process. The process

1_{Gold-toning daguerreotypes is known also as gilding, a process of rapid hardening of the image particles in}

which mercury is replaced by gold. Barger 1991: 159.

2_{Find several examples of processes for the reproduction and transfer of daguerreotypes in Eder: 1927: 35.} 3_{Larsen 1984: 17 and ASTM B832-93 (2013) 2013: 1.}

4_{See for example Meissner, Doktor and Mach on electrotypes from sculptures (2000) and Wharton on}

electrotypes from medals (1984).

5_{As model for daguerreotype served a transmission electron microscopy (TEM) cross-section image. Marquis et}

al. 2015: 441.

(8)

is best understood when comparing an electrotype with its matching daguerreotype. However, of the known plates existing, only four are stored together with their matching daguerreotypes. To date, Jürgens has analysed two daguerreotype electrotypes and their matching daguerreotypes, and one presumed daguerreotype electrotype without a matching daguerreotype. However, the analysis of but two pairs of plates (daguerreotype and electrotype) and one single electrotype plate is statistically insufficient to confirm that these objects are indeed daguerreotype electrotypes. In addition, it is unclear whether the results of the analysis performed so far constitute typical characteristics of daguerreotype electrotypes or if they are unique to the examined plates. One way of achieving a better understanding of the process is to reconstruct the process following historical recipes. This was the main focus of this thesis.

1.2. Research objectives

The main objective was the reconstruction of the electrotype process as applied to daguerreotypes based on historical instructions. Further questions concerned the translation of historical instructions and materials to materials available today, and also the practical parameters for the reconstruction, such as the appropriate consistency of the electrodeposit. The information gained from making reconstructions should:

contribute to the identification of electrotype plates in collections, enhance knowledge on the material-technological aspects of the process, help establish preservation measures for the plates, as far as possible.

The results of this research can eventually lead to the development of a protocol for the identification of electrotypes made from daguerreotypes. However, due to time constraints, the development of such a protocol is beyond the scope of this research.

1.3. Methodology

In order to establish a basis for the reconstruction experiments, a detailed literature study of historical, technical and art historical sources was conducted first (Chapter 2). Nine historical process descriptions from between 1841 and 1860 were found. Of these, the instructions written by Noël M. P. Lerebours in 1843 were chosen as the main source to use for reconstructing daguerreotype electrotypes. Relevant details in other sources that complemented Lerebours’ instructions were added to the reconstruction description as deemed necessary. Modern literature from contemporary practitioners of electrotypy was used to gain a contemporary understanding of the electrochemical principles and translate information from historical sources to materials available today.

An unpublished report by Martin Jürgens (2015) on the detailed examination of a daguerreotype and its matching electrotype (DM-69896 1 & 2) owned by the Deutsches Museum in Munich, Germany, aided particularly in the understanding of process-inherent features of these objects.7_The structure Jürgens used in his report served as an example in the examination of historical plates during this research.

(9)

Three historical plates were examined to obtain a first-hand reference for the comparison of visual features and the results of measurements and analysis (4.1): a daguerreotype (SM-1927-1680 1) and its matching electrotype (SM-1927-1680 2) from the London Science Museum and a single electrotype plate in the private reference collection of Nicholas Burnett, Museum Conservation Services Ltd., (NB plate) were examined together with Martin Jürgens and Nicholas Burnett in Cambridge, UK. The examination of these objects took place in situ within two days. During this relatively short time frame and due to the fact that some analytical instruments were not available, not all analysis could be performed that had been performed by Jürgens on the DM-69898 plates. The observed features of the daguerreotype and the copper plates that are probably a result of the electrotype process are described, and results from the reconstruction are compared to these.

Next, experiments were performed to reconstruct the electrotype process following Lerebours’ instructions. Initial testing was carried out on small silver coupons that simulated the silver surface of a daguerreotype’s bare plate. The parameters that proved to be successful were then used to make electrotypes from modern and historical daguerreotypes. Modern daguerreotypes were produced specifically for the reconstruction and historical daguerreotypes were taken from the University of Amsterdam (UvA) and Rijksmuseum Amsterdam study collections.

The results of the reconstruction and the historical objects were photographically documented and examined visually in ambient light at various angles of incidence and in ultraviolet (UV) fluorescence, as well as with optical stereomicroscopy and digital microscopy. This visual examination had the goal of determining process-specific features of daguerreotype electrotypes. X-ray Fluorescence Spectroscopy (XRF) analysis was conducted in order to detect the elemental composition of the materials involved. The implications of the process of separating the two plates for the microstructure of the plates’ surfaces were of particular interest for their characterization. For this reason, the reconstructions and the historical objects were also examined in a Scanning Electron Microscope with Energy Dispersive Spectroscopy (SEM-EDS). In general, the integrity of the plates was preserved by using only non-invasive examination techniques.

(10)

2. Literature review

2.1. Terminology

The terminology for the process discussed in this thesis varies in historical sources and in different languages (see Appendix I). The terms ‘galvanoplasty’, ‘electroforming’ and ‘electrotyping’ are the most common terms used in historical and contemporary literature. These terms are used in a general sense and not exclusively with regard to daguerreotypes. The chemist William Draper introduced the term ‘tithonotype’ in 1843 when discussing reproduction methods for daguerreotypes. However, this term is also applied to other reproduction processes.9_{In at least one source, the apparatus itself is} referred to as an electrotype.10 As the term ‘electrotype’ is most common in both historical sources as well as in contemporary conservation literature, it is this term that will be used in this thesis to describe the object made by means of electrotypy.11 To refer specifically to the electrotype from a daguerreotype the term ‘daguerreotype electrotype’ will be used. The action of copying daguerreotypes by electrotypy will be referred to as ‘electrotyping’ and the process will be referred to as the ‘electrotype process’.

2.2. The daguerreotype

In short, making a gold-toned daguerreotype entails the polishing of a silver-plated copper plate until the surface is highly reflective, much like that of a mirror. Polishing is usually performed in the direction opposite to the way the image is viewed, for example, a daguerreotype in portrait format is polished horizontally.12 The silver surface is made light-sensitive by exposure to the halogen vapours iodine and/or bromine.After exposure to light from a lens in a camera, the image is developed by fuming the plate with mercury vapours. This results in the creation of image-forming particles on the plate’s surface, which consist of silver-mercury amalgam.13_{Residual, unexposed silver salts are then} removed by fixing the plate in a solution of sodium thiosulfate. From 1840 on, this is followed by gold-toning using a gold chloride solution.14_{Recently, Marquis et al. and Vicenzi, Landin and Herzing} found gold-toned daguerreotypes to have a continuous layer of silver and gold covering the silver surface and the silver-mercury particles. This layer was found to anchor the particles to the substrate by fully encapsulating them instead of only coating them on the outside.15

The daguerreotype thus holds a negative image with a particulate structure. This structure is only perceived as an image as a result of the difference between diffuse reflection of light at the relatively rough surface of the image particles and direct reflection at the smooth polished silver surface. The daguerreotype image appears positive when the silver surface reflects dark contrasting with the

9_{Draper introduced the term tithonotype in 1843. Firstly, he refers to copies of daguerreotypes using isinglass,}

and at a later point he also mentions copper tithonotypes. Draper 1843 (May): 366; 1843 (September): 175.

10_{Lerebours 1843 (September): 117.} 11_{McLeod 2010, Scott 2013.}

12_{Graphics atlas. 2017 Image Permanence Institute. 26 February 2017.}

<http://www.graphicsatlas.org/identification/?process_id=366>.

13_{Barger 1991: 120.} 14_{Barger 1991: 38.}

(11)

whitish-reflecting image particles, and negative when the silver surface reflects light. In the latter case, the image particles block the reflected light and form dark silhouettes.

2.3. Introduction to electrochemical terms

In the following section, a brief outline of the electrochemical principles of the process is provided for better understanding and clarification of some general terms that are used throughout the thesis. Electrotyping or electroforming is commonly considered to be a form of electroplating.16 The main difference between the processes is the following: whereas the adhesion of the deposited coating to the substrate is desirable in electroplating, in electrotyping the substrate is treated to specifically prevent adhesion. However, the electrochemical principles for both processes are basically the same. Figure 2 shows a schematic diagram of the electrotype process.

An electrolytic cell is supplied with an external potential source, which can be a battery or a rectifier. A direct current of electrons passes through a solution of metal ions known as the electrolyte. In the particular case of electrotyping daguerreotypes, the electrolyte is an acidic copper (II) sulfate solution consisting of copper (II) sulfate dissolved in water and sulfuric acid. This solution contains positively charged copper (II) ions (cations), negatively charged sulfate ions (anions) and hydronium cations. The electric current is connected to the electrolyte by means of electrodes. The negatively charged electrode (cathode), in this case the daguerreotype, attracts the copper cations and provides them with electrons. As a result, the copper cations are reduced to metallic copper, and electrodeposition takes place on the surface of the daguerreotype. The reduction reaction applicable to this process is the following:

Cu2+_{+ 2e}-_{Cu (s)}

The hydronium cations are also attracted to the daguerreotype, but, according to the Electromotive Force Series, the reduction potential of the hydronium ion is lower than that of the copper (II) cation.17 This means that the copper (II) is more reactive and that copper will precipitate on the plate first. The

16_{Lowenheim 1978: 115.}

17_{A list of the Standard Electromotive Force Potentials is given for example in Jones 1996: 44.}

Figure 2. Schematic electrolytic cell. Illustration: M. Pilko.

(12)

hydronium only gets reduced if the potential is high enough and if the ratio of hydronium electrons to copper cations is higher.18_{This is usually not the case for electroplating when conventional settings} are used.19 The oxidation reaction takes place at the positive electrode, or anode, which in the case of the daguerreotype electrotype is a plate of copper. The anode closes the electric circuit by removing electrons from the copper. As a result, the copper plate is gradually corroded while continuously supplying the electrolyte with copper cations in order to maintain a sufficient amount of copper cations in the bath. The oxidation reaction applicable to this process is the following:

Cu (s) Cu2+ _{+ 2e}

-Under acidic conditions the sulfate anion can be reduced to sulfurous acid (and water) but this reduction potential is lower than that of the reduction of the copper (II) cation to copper. Moreover, the sulfate anions will migrate towards the positively charged anodic surface of the copper plate, and will not be oxidized unless the supplied direct current is very large.20

For a more in depth understanding of the transport of metal ions through the electrolyte and their adsorption and precipitation at the cathodic surface, see for example Nasser Kanani.21

Current density

When a direct current passes through the cathode and enters the electrolyte, copper will precipitate on the cathode, but the deposit can be of differing quality, ranging from crystalline and non-coherent, dark brown, and brittle, to high quality copper that is coherent, rose-coloured, and flexible. In modern electroplating, the correct electric intensity is calculated in advance. The key parameter is not the total current, but the current density.22_{This is the current flowing to a unit area of an electrode surface, in} Ampere per square decimetre (A/dm2).23 The current density needs to be applied within certain limits to ensure a high-quality deposit. For electrotyping with an acid copper sulfate electrolyte, Larsen advises a cathode current density between 0.5-1.5 A/dm2_.24_{For electrotyping with the same} electrolyte, ASTM B832-93 allows a greater margin of 1-10 A/dm2_.25

2.4. Historical context of daguerreotype electrotypes

Electrotyping dates back to observations made by several individuals of the 19th_{century. Among these} was the Prussian scientist Moritz Hermann von Jacobi, who was the first to understand its practical possibilities in 1838.26_{It remains unclear who the first person was to electrotype a daguerreotype and} when precisely this took place, and it is possible that several individuals experimented with the process independently. In 1843, the optician and daguerreotypist of the observatory in Paris, Noël M.

18_{Larsen 1984: 18.} 19_{Larsen 1984: 35.} 20_{Larsen 1984: 22.} 21_{Kanani 2004: 141ff.} 22_{Larsen 1984: 24.}

23_{ASM International Handbook Committee 1998: 4.} 24_{Larsen 1984: 50.}

25_{ASTM B832 -93 (2013) 2013: 6.} 26_{Heinrich, 1938: 566.}

(13)

P. Lerebours, refers to his collaborator, the physician Hippolyte Louis Fizeau, as being the first.27 The German photographer and scientist, Hermann Krone, who gives a full description of the process in his treatise ‘Die für alle Zeit von praktischem Wert bleibenden photographischen Urmethoden’ (1907) also points to Fizeau as the inventor in 1841.28 The proceedings of the French Academy of Sciences indeed report the presentation of daguerreotype electrotypes made by Fizeau on 24 May 1841.29_A different source from 1848 reclaims the invention for the Parisian optician and daguerreotypist Charles Chevalier.30 In his ‘Encyclopedia of printing, photographic and photomechanical processes’ (1990), Luis Nadeau refers to the tithonotype as “a process for obtaining metallic copies of daguerreotypes by electrotypy discovered by J. W. Draper, of New York“. Nadeau does not indicate a specific date.31

Accounts by for example Chevalier or Lerebours give an understanding of the daguerreotype electrotype – the plate itself – as a separate entity, the final result of the process. However, the source for electrotyping daguerreotypes could also originate in the context of printing. Daguerreotype electrotypes are an offspring from earliest photo history when photography was still rivalling with the graphic arts, which had much experience in producing fine images that could easily be printed and spread to a larger audience.32_{An account from as early as 1840 mentions electrotyping daguerreotypes} in one line with electrotyping engraving plates.33 It appears possible that Fizeau at least initially also intended to use the daguerreotype electrotype as a printing plate. This is supported by a short reference to Fizeau in the extensive works of the Austrian photo-chemist Josef Maria Eder.34_{In 1841, Fizeau,} together with his former teacher Alfred Donné, actually developed a method to pull intaglio prints from etched daguerreotypes. His method also partly entailed the use of electroplating.35_{However, Eder} also states that Fizeau had poor results with printing from daguerreotype electrotypes.

Hippolyte Fizeau, Alphonse Poitevin and Walter Woodbury were pioneering experimental photographers to whom some of the known existing daguerreotype electrotypes are currently attributed. References to at least 13 other makers of daguerreotype electrotypes were found in several sources, indicating that the process was possibly more widely used than previously thought (Appendix II – References in literature to makers of daguerreotype electrotypes).

It is not known when the practice of this technique came to an end, but it can be assumed that this coincided with the decline of the daguerreotype process around 1860, at the latest.36

2.5. Current scientific knowledge on the electrotype process as applied to daguerreotypes

To the knowledge of this author, there are no modern publications on electrotyping daguerreotypes, and a reconstruction of the process following historical recipes has not been attempted to date.

As can be seen from the lack of contemporary publications, these objects have not attracted much attention in cultural institutions. Reference to daguerreotype electrotypes in contemporary sources

27_{Lerebours 1843 (September): 117.} 28_{Krone, 1985: 35.}

29_{Académie des Sciences 1841: 957.} 30_{Pelouze 1848: 713.}

31_{Nadeau 1990: 450.}

32_{More information on printing attempts from daguerreotypes is given in: Bonetti 2014: 30- 43.} 33_{Watt Watt 1840: 344.}

34_{Eder 1905: 73.}

35_{Font-Réaulx 2008: 19, Krone 1985: 40.} 36_{Barger, 1991: 2.}

(14)

were found in ‘French Daguerreotypes’ by Janet Buerger (1989). Buerger was curator of photography at the George Eastman House at the time of publishing the book. She writes a short note on Fizeau’s duplication methods for daguerreotypes and gives an illustration of a daguerreotype and its “photogalvanic copy” (see Appendix III, No. 4).37 Susan Barger, who conducted extensive research on daguerreotypes at the Materials Research Laboratory of the Pennsylvania State University, mentions Draper’s tithonotype shortly with reference to Draper’s original text ‘Note on the Tithonotype’ from 1843, which is one of the recipes considered in this thesis.38 In his book on German daguerreotypes, the German conservator Jochen Voigt briefly mentions a daguerreotype electrotype from the Grassi collection in Leipzig, of which he includes two images.39_{The plate in the Grassi collection is listed} along with three other assumed daguerreotype electrotypes on Daguerreobase, an online daguerreotype database managed by the conservation department of The Netherlands Museum of Photography in Rotterdam.40

In their article ’Electrotypes in Science and Art’, David Scott, a professor in the UCLA conservation program, and Donna Stevens, a senior metals conservator at the Victoria & Albert Museum name Levi Hill as a maker of daguerreotype electrotypes.41_{However, Hill’s process} description is an exact transcript of Lerebour’s text from 1843.42 Upon review, historical sources on electrotyping daguerreotypes appear more numerous than they substantially are and, as in the example of Hill, simply appear to be republished by other authors. In their publication, Scott and Stevens take a close look at the microstructure of electrotype replicas such as vases, which prove to be very different from actual worked and annealed copper artefacts. Even though this kind of research might be considered helpful in the identification of electrotypes in general, the applied method is destructive, so other kinds of analysis such as SEM-EDS appear more suitable for research on daguerreotype electrotypes.

The most substantial research on daguerreotype electrotypes is documented in the unpublished report by Martin Jürgens (see chapter 1.3.) on the DM 69896 plates. It includes results of photo documentation and examination using a 3D digital microscope (Hirox), XRF, SEM-EDS, Infrared Fourier Transform Spectroscopy (FTIR) and UV fluorescence.

The documentation of the edges of the DM-69896 electrotype was particularly useful for the reconstruction performed in this thesis. The copper edges differ from those of the other examined plates in that they display “typical matte, bumpy, nodular growth structures” (Figure 3).43 Particularly informative was also Jürgens’ observation of wax traces along the edges on the recto of the daguerreotype (Figure 4). The presence of wax on the recto could potentially have been relevant in the making of the electrotype.

37_{Buerger 1989: 86.} 38_{Barger 1991: 43.} 39_{Voigt 2004: 42.} 40_{Daguerreobase. 6 March 2017.} <http://www.daguerreobase.org/de/browse/indeling/detail?q_searchfield=galvano&language=de-DE>. 41_{Scott, Stevens 2013: 192.} 42_{Hill 1850: 60-63.} 43_{Jürgens 2015: 8.}

(15)

Figure 3. DM-69896 electrotype verso, raking light.

Photo: M. Jürgens.

Figure 4. DM-69896 daguerreotype, recto, wax along right edge (raking light photomicrograph), scale bar is 800 µm.

Photo: M. Jürgens.

Jürgens’ research with SEM-EDS indicated that approximately half of all image particles were transferred from the daguerreotype to the electrotype, leaving craters on the surface of the daguerreotype. It is initially surprising that, despite retaining a surface damaged from electrotyping, the daguerreotype still displays a fine and contrasty image. Jürgens remarked that diffuse reflection of light at the craters must be enough to make the image very visible.44 This implies that a protruding image particle, a crater or even a pit on the electrotype is apparently able to scatter incident light in a similar way, as illustrated in Figure 5.

Figure 5.Schematic reflection of light at the daguerreotype and electrotype surface.

Incident light is specularly reflected at the smooth daguerreotype and electrotype surface. Disruptions of the smooth surface by particles, craters or pits result in diffuse reflection.

Illustration: M. Pilko.

2.6. Instructions on electrotyping daguerreotypes in 19

th

century sources

2.6.1. Choice of historical process description

We have no information on the production of the historical daguerreotype electrotypes found in collections today. Therefore, the practical instructions on electrotyping daguerreotypes found in

(16)

treatises from nine authors, dating from the daguerrean era, were very helpful.45 The treatises vary in length from half a page to full eight pages (see Appendix IV – Historical instructions on electrotyping daguerreotypes). The different instructions also vary in setup and degree of detail but they basically provide similar recipes and tend to be repetitive to a large extent. This coincides with the fact that the underlying chemical principles are essentially the same (see 2.3. Introduction to electrochemical terms).

It is difficult to assess to which extent an author actually worked with the process himself or rather just copied other accounts. For example, the metallurgist Alfred Smee states that he has not practiced the process himself, but that he is instead reporting an account of Dr. Symon and Mr. Horn.46_{In his} treatise, Humphrey Davy quotes a description of the process by daguerreotypist John Fitzgibbon. A written record on electrotyping daguerreotypes actually written by one of the persons to whom historical objects are currently attributed could not be found. The authors appear to have at least partly copied texts from one another. For example, Lerebours compares the appropriate thickness of the electrotype plate to that of a stout card, and the exact same comparison is made by von Pauly.47_Given these circumstances, this author considered the information in the earliest sources to be the most original account on making daguerreotype electrotypes. In addition, this author focused on authors who actually practiced the process themselves. For instance, Chevalier declares that he practiced the process himself, and he gives the first account of electrotyping daguerreotypes as early as 1841. However, the reconstruction of his setup (Figure 6) confronted this author with practical constraints: the daguerreotype to be reproduced is integrated in a battery circuit that requires the use of mercury. As mercury is extremely poisonous, this author decided to use another historical description as starting point.

Figure 6. Illustration of Chevalier’s assumed setup for

electrotyping daguerreotypes. Chevalier used a device made by Tito Puliti from Florence, which essentially consisted of two vessels, the inner being porous and containing a copper sulfate solution and a cathode, and the outer containing diluted sulfuric acid and an anode (Chevalier 1840: 61). This device appears similar to a horizontal version of a single-cell apparatus as described, for example, in historic manuals by the electrical engineer Walker Charles and the metallurgist Alfred Smee (Smee 1843: 59, Walker 1848: 34). Illustration: M. Pilko.

The second earliest account is from a rivalling optician, Lerebours, who, in his manual ‘Traité de photographie derniers perfectionnements apportés au daguerréotype’ from 1843, includes a chapter on copying daguerreotypes by the electrotype process. As the original French manual was translated into English in the same year, Lerebours’ instructions may have reached many readers. Lerebours describes a setup with an external power supply, for which a battery was used (called a Bunsen element).48 This

45_{Authors considered were: Charles Louis Chevalier (1841), Noël M. P. Lerebours (1843), Theodor von Pauly}

(1843), John William Draper (1843), George Shaw (1844), L. E. Uhlenhut (1849), Alfred Smee (1841/1852), Samuel Humphrey (1858 and James Napier (1860). After completion of the reconstruction work a tenth source by A. Lipowitz (1845) was found, which for reasons of time could not be considered any further.

46_{Smee 1841: 134-135; Smee 1852: 328.} 47_{Lerebours 1843: 118; von Pauly 1843: 83.} 48_{Lerebours 1843: 117.}

(17)

setup can be reconstructed to a large extent and can be imagined to work similarly to the illustration in Figure 2, chapter 2.3. Due to the fact that they both appear relatively original, influential and feasible to follow, Lerebours’ instructions were chosen as the main source for the reconstruction part of this study. However, no clear evidence could be found as to whether Lerebours actually practiced the process himself.

2.6.2. Lerebours' process description supplemented by other sources as implemented in initial testing and the reconstruction

Lerebours’ setup consists of an external battery and a glass basin with a saturated solution of copper sulfate (see for the original instructions Appendix IV, No. 3 and No. 4). The solution is considered to be sufficiently saturated when copper sulfate crystals cease to dissolve after shaking the solution. According to Lerebours, the following basic steps were performed:

1. The copper plate (positive electrode) is connected to the negative pole of the battery (which is carbon). The gold-toned daguerreotype is connected via a wire, attached at one corner, to the positive pole of the battery (which is zinc).

2. The verso and edges of the daguerreotype are coated with melted beeswax or beeswax in

turpentine (2:1) in order to avoid unnecessary deposit of copper. The wax layer requires a certain thickness that is not sufficiently defined by Lerebours. The spot at which the wire is attached to the daguerreotype is kept free of wax in order to allow current to flow.

3. The copper plate is immersed first in the glass basin holding the copper sulfate solution. 4. The daguerreotype is then immersed in the bath. Copper will immediately start to deposit on the

daguerreotype and the copper plate will dissolve. During the process, the daguerreotype should be removed only for a short time, if at all, as oxidation may occur and interrupt the process.

5. When the deposit is sufficiently thick (as thick as a “stout card”), the daguerreotype with copper deposition is removed from the bath and rinsed in water. It is then dried with blotters or sawdust. Droplets are not allowed to remain as these could penetrate between the plates and stain them. Drying may be sped up by wetting the surface with spirits of wine to retain the pearly colour of the copper.

6. The plate of copper deposition is pried off the daguerreotype. If necessary, the daguerreotype is cut down by 2 mm at all edges in order to be able to separate the new copper sheet from it.

7. Touching the electrotype is to be avoided, and since copper will oxidize, the electrotype should immediately be sealed in a frame.

Lerebours’ instructions were partly not clear to this author or lacked information necessary to reconstruct the process. In the following, Lerebours’ instructions were therefore supplemented with information from other historical and modern sources as well as translated to modern materials where this was regarded necessary:

Setup of the electrolytic cell

Lerebours does not specify the position of the electrodes, but the metallurgist Alfred Smee suggests both horizontal and vertical setups for the general purpose of electrotyping that could both be

(18)

appropriate for daguerreotypes (Figure 7, Figure 8).50 According to Smee, the vertical setup is more suitable for slow plating and for small plates.51 _{The vertical setup was used in the reconstruction} experiments as it also allowed for a simple fixture for the electrodes.

Figure 7 (left). Vertical setup for electrotyping, with a battery to the right side.52

Figure 8 (right). Horizontal setup for electrotyping, with a battery to the right side.53

Connecting the electrodes

Lerebours’ description on connecting the electrodes with the battery is contradictory to the contemporary understanding of a functioning electrolytic cell. Even though Lerebours' descriptions are in accordance with historical accounts by Smee and Shaw, modern terminology is different.54 As explained in Section 2.3, in an electrolytic cell the copper plate is termed the positive electrode (anode) and connected to the plus pole of the battery whereas the daguerreotype is the negative electrode (cathode) and connected to the minus pole of the battery.55

Mixing the copper sulfate solution

In contemporary electrotyping, sulfuric acid is generally added to the copper sulfate solution to improve the quality of the copper deposit. For this reason, it was used in all the reconstructions even though Lerebours does not mention it. This was justified by another source from 1840, which has detailed instructions on the preparation of the electrolyte, which also mention the addition of sulfuric acid.56 Anton Martin also describes the use of sulfuric acid for electrotyping and attributes to it an increased flexibility of the copper deposit.57_{For the preparation of the saturated solution, the ratio of} sulfuric acid, copper sulfate and water was used as described in Larsen.58 Other additions to the electrolyte, which are common in contemporary electrotyping and which are also described by Larsen, were disregarded, as no reference to their use could be found in 19th_{century literature. Future research} might study the impact of different electrolytic mixtures on making daguerreotype electrotypes.

51_{Smee 1851: 294-295.} 52_{Smee 1851: 99.} 53_{Smee 1851: 107.} 54_{Smee 1851: 98, Shaw 1844: 175.} 55_{Larsen 1984: 42.}

56_{Allgemeinfaßliche Beschreibung des Verfahrens zur Herstellung galvanischer Kupferstiche 1840: 5-9.} 57_{Martin 1856: 186, 188.}

(19)

The anodic copper plate

Krone states that a cleaned copper plate having at least the size of the daguerreotype should be used as an anode and that this plate will be consumed during the electrotyping process.59 From this it was concluded that any oxidation products present on the copper plate would need to be removed first and that the copper plate might need to be replaced regularly.

Fixture of electrodes

Various methods to fix the battery wire to the electrodes in order to suspend them in the electrolytic bath are known, and each method would probably have a different physical impact on the daguerreotype. Lerebours mentions binding screws with which the battery wire is attached to the daguerreotype.60_{Historical binding screws were not available and the historical battery wire probably} also looked different. Paper clips were chosen instead of the binding screws to allow the current to flow.

Coating of the daguerreotype

According to Lerebours, the daguerreotype is partially coated with wax to prevent electrodeposition where it is unwanted. Pure, hot wax is difficult to apply in a controlled way, due to its rapid cooling after which it becomes thick and difficult to spread. Even though diluting the wax with white spirits facilitates its application, a relatively thick wax layer requires a long time to harden due to the long evaporation time of the solvent. The use of Paraloid B72 in a solvent instead of wax allowed for a more controlled application of the protective coating with a much shorter drying time.

Distance between electrodes

The distance between the cathode and the anode is not specified by Lerebours. However, the metallurgist Moritz von Jacobi indicates that the electrodes should not be placed closer than 1½ inches to 2 inches.61_{Another metallurgist, Partridge, states that “the moulds should be separated from the} anodes by a distance of about two inches; but if it is found that the deposit is very dark in colour or granulated in texture, this distance may be increased, thereby increasing the resistance of the solution, which is equivalent in its effect to cutting down the current strength.”62_{This indicates that there is a} certain margin of error for the right distance. In the reconstruction of the process, a distance was therefore chosen that fitted the dimensions of the available basin and the required plating intensity.

Current intensity

A single bunsen battery, which Lerebours recommends, generates a potential between 1.86 V to 1.96 V.63_{For this study, a low power rectifier was chosen, since it can simulate the characteristics of a} Bunsen cell, the reconstruction of which would have been beyond the scope of this research. Von Pauly judges the strength of the electric current by the amount of hydrogen bubbles. If the electric current is too strong, hydrogen bubbles will rapidly form at the daguerreotype, and the deposited

59_{Krone and Schmidt 1907: 35.}

60_{Lerebours 1843: 119, Chevalier mentions soldering as an alternative. Chevalier 1841: 62.} 61_{Jacobi 1840: 43.}

62_{Partridge 1908: 102.} 63_{Sivasankar 2008: 129.}

(20)

copper will be spongy and weak.64 Smee writes that “the metals are invariably thrown down as a black powder, when the current of electricity is so strong in relation to the strength of the solution, that hydrogen is violently evolved from the negative plate of the decomposition cell.”65 Humphrey explicitly warns of a too strong electric current.66

Duration of electrotyping

Different sources indicate greatly varying durations for electrotyping daguerreotypes, ranging between 3 to 20 hours. Lerebours states that a single Bunsen battery reproduces a 16 x 22 cm plate within just a few hours. According to von Pauly, Gaudin also succeeded in reproducing daguerreotypes within only 3 hours.67 Draper, on the other hand, indicates that the complete process takes 12-20 hours.68 Von Pauly suggests starting with a cold bath and warming up the solution to 30-40°C once the daguerreotype has received an initial coating of copper, which accelerates plating speed.69_This accelerated plating would then still take 8-12 hours.70 From these diverse specifications it was concluded that, firstly, there is some flexibility in the duration of the process that will result in different thicknesses of the electrotype, and secondly, that the precise duration appears to be dependent on whether the plating speed was accelerated at a certain point or not. For the reconstruction the plating speed was held steady throughout the whole process as this provided the most helpful insight into the characteristics of the copper growth. The duration of electrotyping was evaluted according to the desired thickness of the electrodeposit. Lerebours’ specification of a “stout card” was determined to be approximately 0.4 mm.71

Method of separation

The historical literature contains several methods of separating the electrotype from the daguerreotype. In regards to Lerebours' recommendations on cutting the edges of the plate, Krone remarks that if the daguerreotype is to be electrotyped more than once, then cutting the edges is to be avoided in order to prevent it from getting smaller each time. Chevalier describes filing off the copper rims that form along the edges and then “a knife blade is carefully inserted between the two metallic plates and towards the corners, gently raising and lowering it, to push the plates apart.”72 During the initial electrotyping tests, Lerebours’ instructions were followed, however the edges of the plates were not trimmed in the actual reconstructions. Trimming the edges would have further reduced the size of the already small plates.

Number of copies

According to von Pauly and Napier, several daguerreotype electrotypes can be made from a single daguerreotype.73_{Fitzgibbon specifies that he made 12 copies from one daguerreotype.}74_Draper

64_{von Pauly 1843: 84.} 65_{Smee 1841: 66.} 66_{Humphrey 1858: 168.} 67_{Von Pauly 1843: 84.} 68_{Draper 1843: 175.} 69_{von Pauly 1843: 82.} 70_{Humphrey 1858: 168.} 71_{Lerebours 1843: 118.} 72_{Chevalier 1841: 64.}

(21)

remarks that more than one copy can be obtained by electrotyping the electrotype.75 This indicates that if the reconstruction is correctly performed, it should be possible to make more than one reproduction. For this reason, making a second reproduction was attempted in the reconstruction.

Considerations in the making of daguerreotypes used for electrotyping

A crucial part in the reconstruction concerned the daguerreotype itself. For technical and ethical reasons, the use of historical daguerreotypes was initially not intended in this research. Firstly, the number of historical daguerreotypes is limited. Conservators regard each daguerreotype as unique cultural heritage that should be preserved. Secondly, historical daguerreotypes have aged naturally for over 170 years, their conservation history is not well documented, and the aging of daguerreotypes is not yet fully understood. For this reason, historical daguerreotypes are not representative of the daguerreotypes that were electrotyped at the period of production in the 19th_{century. Therefore, it was} initially decided to use modern daguerreotypes, made according to Daguerre’s original process, and prepared specifically for this reconstruction experiment. A study indicates that modern daguerreotypes can serve as “a good representation for historic plates” even though, as with any obsolete photographic process, practice and materials may have undergone change.76 Furthermore, modern daguerreotypes have the advantage that the parameters of their making and their subsequent history is known. However, during this research, the reconstruction plates produced unexpected results, and it was deemed necessary to use historical daguerreotypes as control for a number of aspects of the reconstruction.

The contemporary daguerreotypist Anton Orlov was commissioned by this author to make five daguerreotypes (designated ‘AO plates’). Given the time of the year during which this research was carried out, it would have been difficult to make daguerreotypes in Europe, as UV and sun light are weak in winter and daguerreotype plates are most sensitive to UV light and the blue region of the spectrum.77 However, Orlov had appropriate climate conditions in San Diego, USA, were he lives. A small plate size (a ninth plate, 72 x 54 mm), similar to the size of the DM-69896 plates, was chosen.78 A large contrast range, sharpness and recognizability were criteria in the selection of the image, all of which are sufficiently present in the view to a backyard as seen from Orlov’s studio window.

Historical instructions specifically for the production of daguerreotypes to be electrotyped are provided by Samuel Humphrey, and Orlov followed these as closely as possible. Important points were that more iodine and less accelerators were used and the development using mercury lasted ideally relatively long, between 6 to 8 minutes.79_{Accelerators are additional halogen vapours, in} particular bromine, which are used in increasing the light sensitivity of the silver plate.80_Gold-toning was to be performed slowly, and at low heat, in order to achieve a uniform result. For the reconstructions, it was also important that after gold-toning, the daguerreotype be “kept for a day or two, so it may become enfilmed with air.”81 This recommendation suggests that corrosion products

74_{Humphrey 1858: 169.}

75_{Draper 1843 (September): 176.}

76_{Ravines, Chan, Nowell, McElroy 2013: 158.} 77_{Barger 1991: 30.}

78_{Plate sizes of daguerreotypes are traditionally specified in proportion to a 1/1 or whole plate, which is 216 x}

162 mm (8 x 6 pied). Accordingly, a ninth plate is 72 x 54 mm. Eder 1927: 12.

79_{Humphrey 1858: 169.} 80_{Barger 1991: 45.}

(22)

may benefit the electrotype process. A statement by Smee actually indicates that “a film of oxide at other times a thin film of sulphuret” will prevent adhesion between the metals.82

(23)

3. Experimental Procedure

3.1. Initial testing and reconstructions

This section describes initial testing and the attempted reconstructions. The main goal in the available time was to put the process in practice, so the many variables described in the sections above could be tested only to a minor extent.

The setup of the electrolytic cell and the mixing of the electrolyte used for the tests is described first, followed by details of initial testing and producing the reconstructions.

3.1.1. Setup of the electrolytic cell

Figure 9. Setup of the electrolytic cell for the reconstructions. Letters are defined in the text below. Photo: M. Pilko.

Figure 10. DB

daguerreotype suspended in the electrolytic bath. Photo: M. Pilko.

The experimental setup is shown in Figure 9. The setup consisted of a glass basin (A), 18 x 18 x 13 cm (H x W x D) in which the specimen to be electrotyped (B) was suspended vertically. A Weir 460 rectifier with a DC of 0-60V and a maximum of 3A was used (C). The glass basin was filled with an acidic copper (II) sulfate solution. Two copper wires (diameter = 1 mm) were placed over the shorter sides of the glass basin at a distance of approximately 8 cm (D). A copper plate (anode) (E) cut to a size of 50 x 65 x 0.5 mm with a drilled hole at each corner of one of the long sides was used as anode.83 The copper plate was replaced with each test and rubbed with steel wool prior to its use in order to remove copper oxidation. It was then suspended in the electrolytic bath held by wires hooked into the drilled holes. The rectifier was switched on and amperage and potential were adjusted to the desired values. Next, a specimen (cathode) (B) was suspended on the other wire facing the copper

83_{During initial testing, the relatively large size of the anode did not visibly influence the result. However, it is}

generally advised to have both the anode and the cathode at approximately the same size, which was considered for the actual reconstruction.

(24)

plate. Initially, specimens were suspended in the electrolytic bath by wires hooked into previously drilled holes, later they were fixed with metallic paper clamps (Figure 10). With few exceptions, both electrodes were partially immersed so that the wires would not touch the electrolyte. This made it possible to retain an unprocessed area as a control, and it also made it unnecessary to isolate the copper wires. A cable with battery clamps (F) connected the cathode to the minus pole (G) of the rectifier (black input). A second cable connected the anode to the plus pole (H) of the rectifier (red input). A sheet of polyester (I) on top of the basin weighed down with cardboard was used to prevent evaporation of the electrolyte.

Mixing the electrolyte

The acidic copper electrolyte consisted of 300 g copper (II) sulfate in 1500 ml demineralised water and 75 g sulfuric acid (95%, D = 1.84 g/cm3_{). The copper (II) sulfate was added to boiling water,} resulting in a chemical reaction that causes the solution to briefly foam up. A beaker considerably larger than the required size was therefore used in order to prevent spillage. After the solution cooled down, 75 g sulfuric acid was added while stirring. The electrolyte was used at room temperature (20°C). Measuring the electrolyte with pH strips (pH 0-14 universal indicator by MColorpHast) yielded an indicative value of 1 pH. The saturation of the solution was maintained by regularly adding copper sulfate until the crystals ceased to dissolve.

3.1.2. Initial tests with silver coupons

Initial testing with silver coupons had the goal of determining working parameters for: (I) the appropriate consistency of the copper deposit by testing different current densities; (II) the duration of electrotyping needed to obtain a deposit with an approximate thickness of 0.4 mm; and (III) the feasibility of separating the electrodeposit from the coupon.

(I) Current densities were tested in a range from 25 to 2.5 A/dm2_.

(II) The thickness of the electrodeposit and the time taken to obtain it was mainly visually assessed, but it could be mathematically calculated as well.

(III) Rims of copper that formed around the edges of the coupon were cut off.

Description of the cathodes used in initial testing

The coupons used were obtained from a 925 sterling, polished silver plate, cut into 20 x 20 x 0.5 mm or 10 x 10 x 0.5 mm (H x W x D) squares. A 2 mm hole was drilled through the coupon at the top centre. The coupons were subsequently polished with three different grades of sanding paper (500, 1,200, 1,600 grits) to remove deep scratches. A mirror like, high polish finish was provided with a polishing machine using the polishing paste Rouge de Paris. The polishing paste was cleaned off with white spirit and the coupons were degreased with analytical grade ethanol. A 40% solution of Paraloid B72 in analytical grade acetone was applied on the verso and the edges to prevent copper deposition. The specimens were left to rest for several days to let oxidation occur as indicated in historical instructions.

(25)

3.1.3. Reconstruction tests with daguerreotypes

Successfully established parameters from initial testing were then applied to daguerreotypes with the objective of obtaining one or more electrotypes that would show a reversed image of the daguerreotype, with a positive image in reflected dark and a negative image in reflected light.

Description of the cathodes used in the reconstruction tests

The main test specimens were four gold-toned daguerreotypes made by the contemporary daguerreotypist Anton Orlov (AO-1 to AO-4 plates). The plate dimensions varied slightly at approximately 60 x 48 x 0.5 mm (H x W x D). Orlov used daguerreotype substrates made by the traditional roll clad process.84_{The plates were polished with a Makita electric random sander, using} red rouge and lamp black as a polishing compound. The plates were sensitized consecutively with iodine for 30-40 seconds, bromine for 8-12 seconds, and finally iodine again for 8 seconds. Light exposure was between 25-30 seconds in a camera fitted with a Schneider Super Angulon 65 mm lens. Processing was performed with mercury fumes at 65°C for 5 minutes, and the plate was fixed in a 2-3% w/v sodium thiosulfate pentahydrate solution in tap water. Gold-toning the developed and fixed image was performed by holding the plate on a layer of aluminium foil over a butane torch. The gold-toning solution consisted of a 0.2% w/v solution of gold chloride in distilled water mixed with a solution of 1 % w/v sodium thiosulfate in distilled water. The pH of the solution was buffered by a 2% solution of sodium metaborate: 1 drop per 2 ml of gilding solution.85_{See Appendix VI.1. Anton Orlov} daguerreotypes (AO) for details on each plate.

Other test specimens used as control group included two historical and one modern daguerreotype: A 19th_{century daguerreotype from the UvA study collection, in the following named ‘UvA plate’.} A 19th_{century daguerreotype donated to the Rijksmuseum Amsterdam by Daniel Blau, in the} following ‘DB plate’. The DB plate exhibits scratches overall and appears to have been previously cleaned.

A contemporary daguerreotype by a different maker, Marinus Ortelee, in the following the ‘MO plate’. In contrast to the AO plates, the MO plate has an electroplated substrate. The MO daguerreotype is also only partially gold-toned. It rested for over one year during which the surface likely oxidized to a great extent.

To expand testing possibilities, several plates were cut into pieces using a jeweller’s saw with saw blades Nr 2/0. All daguerreotypes were coated on the verso and the edges with a 40% solution of Paraloid B72 in acetone (Figure 11) to prevent copper deposition. In some cases this coating was also applied along the edges on the recto to prevent copper deposition from forming around the edges. After removal from the electrolytic bath, a knife blade was inserted between the metal layers and the electrodeposit was then pried off by hand.

84_{The earliest substrate used for daguerreotypes are roll-clad plates, made by hammering or rolling a thin silver}

plate on a thicker copper plate. Since the mid-1840s electroplated copper plates became more common. Krone and Schmidt 1907: 10.

85_{The addition of sodium metaborate is not known in Daguerre’s original process, but Orlov commonly uses it}

for gold-toning. Da Silva et al. (2010: 656) also mentions it in the making of modern daguerreotypes for research purposes.

(26)

Figure 11. Coating the edges of the AO-2 daguerreotype with Paraloid B-72 in acetone. Photo: M. Pilko.

Reconstruction tests were carried out either (I) in accordance with Lerebours' instructions as supplemented by other sources, (II) included the application of a coating not mentioned by Lerebours, or (III) attempted to achieve a second reproduction from one daguerreotype. An overview on all tests is given Table 1 and in Appendix VI – Documentation of specimens before and after electrotyping.

(I) The following daguerreotypes were electrotyped according to Lerebours’ instructions as supplemented by other sources and as specified above (2.6.2 and 3.1.1).

AO-3 and AO-4 daguerreotypes. Following the first results, further current densities were tested to see if this parameter influenced the separation of the electrodeposit from the daguerreotype. The goal was an electrodeposit with individual nodular formations in an otherwise fine-grained copper surface on the verso, comparable to the verso of the DM-69896 electrotype and SM-1927-1680 electrotype (see Figure 3, Figure 25). DB and MO daguerreotype. The historical DB plate and the contemporary MO plate

were electrotyped to understand whether consistent results could be achieved with very different daguerreotypes. Electrotyping the partly toned MO plate offered additional insight into the effect of gold-toning on the process.

(II) Following the results from (I), an additional step was introduced to the process, which is not listed in any of the historical instructions: a thin coating was applied to the recto of the daguerreotype prior to electrotyping. The aim was to add a separation layer at which the copper deposit would easily separate, thereby preventing damage to the surface of the daguerreotype. Other parameters remained unchanged. The use of a coating was tested on the following daguerreotypes:

AO-2 daguerreotype. The daguerreotype was coated by swiping over the recto with a piece of deerskin that had previously been dipped into beeswax dissolved in white spirit (2:1). Excess wax was then rubbed off.

UvA daguerreotype. The same wax solution was applied in one single swipe, but excess wax was not removed in an attempt to prevent physical damage to the daguerreotype surface. For the purpose of comparison, the daguerreotype was divided into different sections: the upper area of the plate was not electrotyped, the lower left half was coated with beeswax, and the lower right half was left uncoated.

Electrotyping Daguerreotypes: Reconstruction of an Early Reproduction Technique

Conservation and Restoration programme, Master’s, Photograph Specialisation